MEASURING PREVALENCE OF D&A DURING CHILDBIRTH IN KENYA: THE HESHIMA PROJECT Timothy Abuya, Charity Ndwiga, Lucy Kanya, George Odhiambo, Alice Maranga and.

MEASURING PREVALENCE OF D&A DURING CHILDBIRTH IN KENYA: THE HESHIMA PROJECT Timothy Abuya, Charity Ndwiga, Lucy Kanya, George Odhiambo, Alice Maranga and Charlotte Warren

Respectful Maternity Care Seminar: 24th June 2014George Washington University School of Public Health

Heshima Project objectives

1. Specify types and prevalence of D&A

2. Develop and validate tools for assessing D&A

3. Identify potential drivers of D&A

4. Design, implement and evaluate interventions to reduce D&A

5. Generate lessons for replication and scale up

Why measure the prevalence?•  Is the measurement aimed at developing

interventions/policy response at both global and local levels?– Developing interventions to improve quality of care– For advocacy purposes to bring forth human right

issues of equity/access to quality health care – Improve health indicators (maternal indicators)

• Is it aimed at developing tools that are sensitive enough to help point the magnitude of the problem (academic purpose) ?

How purpose influences measurement approach

Purpose Things to consider for measurement

developing interventions/policy response at both global and local levels?

Multiple sources of data that are cost effective

-take account of client’s perspective

developing tools that are sensitive enough to help point the magnitude of the problem

-Rigorous and multiple sources of data to triangulate /sensitivity of the measurement tools

-Huge sample size in multiple places to account for variations in context

-standardization of tools-language and whether to use single or multiple item response

Defining disrespect and abuse in facility-based childbirth

Structural level:What women and providers

consider poor care, but is caused by system deficiencies

Deviations from national standards of good quality care

Deviations from human rights standards (available, accessible,

acceptable, quality)

Individual level:Normalized D&A:

What women experience as D&A but providers consider normal

When providers are disrespectful and abusive but women consider

it normal

Individual level: actions that all agree are D&A

Initial intervention target

Prevalence Measure

Policy Advocacy

Structural level:System deficiencies that lead to poor care that is accepted and

normalized

Individual level

Structural level

Policy Level 

Multiple data sources for different needs

Purpose Data sources

developing interventions/policy response at both global and local levels

Client provider observations

client exit interviews (Immediate/follow up)

Inventories

Provider/community views

developing tools that are sensitive enough to help point the magnitude of the problem

Client provider observations

client exit interviews (Immediate/follow up)

Community survey

Provider/community views

Case narratives

From the bull’ eye to the actual measurement

Category Experienced by clients

Non confidential care

Treated in a way that violated privacy and/or confidentiality

Non-dignified care

Provider said/used a facial expression that made you feel uncomfortable

Neglect/abandonment

Left unattended when needed help Requests for pain relief ignored

Non-consented care

Treatment given without permission

Physically abused

Slap, pinch, push, beat, poke

Inappropriate demands for payment

Detention for failure to payRequest for a bribe for services

Any occurrence of Disrespect and Abuse

Main question: at any point during labor and delivery were you treated in a way that made you feel humiliated or disrespected

• 20% clients (n= 641) reported they were made to feel humiliated or disrespected at some point during labour and delivery

Type of D&A Experienced by Postpartum Women

Comparison of occurrence of any D&A with socio economic status

% reporting Lowest 20% Others P value

Any D&A

24.2% 19.1% 0.197

Lowest 40% Other categories

21.9% 19.0% 0.367

Lowest 60% Other categories

20.6% 19.5% 0.729

Highest 20% Other categories

15.5% 21.3% 0.143

Comparison of highest and lowest quintiles and D&A experience

Factors associated with D&A

Characteristic Any D&A DetainedRequest for

bribe Parity: ref: no previous children

Between 1-3 children

1.2(0.59,2.3)p=0.621

3.5 (2.2, 5.9)p<0.001*

4.5 (1.2,17.4) p=0.028*

Between 4-9 children

0.9 (0.13,6.8)p=0.984

12.4 (3.2 47.4) p<0.001*

49.4 (8.6,279) p<0.001*

Marital Status (ref: Never Married & Separated) 

Currently Married

0.7 (0.42,1.0)p=0.067

0.2 (0.1,0.34)p<0.001*

0.2 (0.1,0.72)p=0.014*

Age: (ref below 19 years)

20-29 years1.1 0.61,1.8)

p=0.8180.4 (0.12,1.4)P=0.186

0.4 (0.092,1.7) p=0.223

Over 30 years NA0.3 (0.03,1.8)

p=0.185 NA

Prevalence of D & A observed during admission and delivery

Methodological issues to measurement

• What is the best methodological approach to assess the prevalence of D&A?

- Different methods yield different estimates of prevalence of D&A?

-Method for assessing the prevalence depends on purpose

- Key considerations are likely to be- logistical issues of collecting data, cost implications versus propose (policy change/interventions )

Other attempts made to validate the tools

Study Sites and magnitude

Validation of Skilled birth attendance tool at facility and community follow up

Two sites in Kenya-Kiambu and Kisumu (nearly 600 women observed, interviewed on exit and will be followed up this year)

Assessing the impact of voucher on reproductive health behaviors

Over 3000 women interviewed during end line community survey across five counties

Base line community survey of a Kenya Signature program

About 3000 women interviewed in Bungoma County

Lessons learnt • Choice of data collection procedures linked to

purposes-costly and cumbersome (standardizing approach/checklist, level of sensitization of data collectors)

• Contextualizing certain actions/observable behavior into categories that are measurable is difficult

• Some aspect of D&A are largely considered quality of care issues- but not written in standards of care guidelines

• Measurement of various categories of D&A appears to occur at various birthing stages thus difficult to have one single measure of prevalence

The Population Council conducts research and delivers solutions that improve lives around the world. Big ideas supported by evidence: It’s our model for global change.

Ideas. Evidence. Impact.